论文部分内容阅读
本文从聚类分析的基本思想和实际要求出发,在对现有模糊聚类法的基本特点及适用条件进行深入分析的基础上,提出了一种新的模糊聚类分析方法——综合模糊聚类法。这种方法与其它方法的主要区别在于它建立了确定聚类样本间相似数r_il的合理计算公式。分析及验证结果表明、本文提出的方法对于处理大型复杂系统的聚类问题具有更为广泛的适应性和良好的可靠性。
Based on the basic ideas and practical requirements of cluster analysis, this paper presents a new method of fuzzy clustering analysis based on the analysis of the basic features and applicable conditions of the existing fuzzy clustering method - Class method. The main difference between this method and other methods is that it establishes a reasonable formula for determining the similarity r_il between clustering samples. The analysis and verification results show that the proposed method has more extensive adaptability and good reliability for dealing with the clustering problems of large and complex systems.